The general objective of the proposed research is to fine-map selected T1D loci harboring common risk variants and to determine their mechanism of action and contribution to [unreadable]-cell destruction. In Aim 1a, we propose to discover all variants within the associated linkage disequilibrium (LD) blocks of a frequency high enough to account for the observed association at these loci. We have selected eight loci all of which involve immune mechanisms, all of which have been well established and replicated. Three hundred cases will be re- sequenced for all the loci, selected for carrying the susceptibility alleles in as many as possible;the controls will come from the 1000 genome project. In Aim 1b, we propose to evaluate all variants within the LD blocks for independent effects on T1D risk. Selected tag markers will be genotyped on our collection of 12,000 DNA samples (4,000 cases and both their parents) as well as in a smaller cohort of T1D patients (200) and controls (400) of African-American ancestry. The results will be analyzed by conditional regression to identify polymorphisms whose association with T1D cannot be accounted for by LD to another effect within the block. In Aim 2a, we propose a large-scale, high throughput evaluation of allelic effects, in cis, of the T1D-associated polymorphisms on transcription, splicing and translation. In Aim 2b, we will determine the cellular phenotypes encoded by the T1D-associated variants using peripheral blood cells from cases and healthy controls. For genes with known function, such as IFIH1, PTPN22, RASGRP1, BACH2, UBASH3, this function will be compared, ex vivo, between cells from individuals of different genotypes. For genes whose function is not known, such as CLEC16A, the first step will be to define its function by over-expression and SiRNA knockdown experiments;In Aim 2c, we will use animal models of human loci as an important source of clues to function. We also propose to create knock-in (KI) mice carrying the two alleles of the human polymorphism, starting with the well-mapped loci involving coding variants in the PTPN22 and IFIH1 genes and proceeding to the CLEC16A mose. The mice will be crossed to the NOD background and the effect of the human polymorphism on diabetes susceptibility and on specific aspects of immune dysregulation will be studied. In Aim 2d, we will evaluate epigenetic mechanisms through the analysis of methylation patterns within regulatory regions of genes with expression patterns that correlate with T1D risk. In Aim3a, we will perform detailed phenotype- genotype correlations studies of all loci re-sequenced in search for allele-specific genotype effects. In Aim 3b, we will establish a predictive test using SNP markers that associate most strongly with T1D with sensitivity and specificity that is of clinical relevance in predicting T1D risk. We anticipate these studies will markedly enhance our understanding of the pathogenesis of T1D. PUBLIC HEALTH RELEVANCE: PROJECT NARRATIVE Type 1 Diabetes (T1D) is a multifactorial childhood disease with a strong genetic component. It is one of the most common severe chronic childhood diseases, with life expectancy severely shortened and quality of life compromised. It is caused by autoimmune destruction of the pancreatic beta cells with resulting lack of production of insulin. Although we and other groups have identified several T1D loci through recent genome- wide association studies, the actual causal mechanisms by which these loci confer risk remain unknown. This project proposes to 1) discover all variants within the loci that may account for the observed association and determine their independent effects on T1D risk;2) functionally characterize the variants;and 3) further characterize the etiological pathways through comprehensive genotype/phenotype correlations and establish a proof of concept for use of these variants in risk stratification and future therapeutic guidance (ie, personalized medicine).